EEG Signals Classification for Epileptic Seizure Detection

Abstract

The study introduced in this thesis, presents the application of various approaches for the automatic classification of electroencephalography (EEG) signals, to detect epileptic from normal persons. Our methodology involved employing two distinct classification methods. The first bases on the support vector machine (SVM), while the second uses the convolutional neural network (CNN) combined with bidirectional long short-term memory (Bi-LSTM). The evaluation results showed the superiority of the second method, as the accuracy of classifying the various epileptic and normal cases reached 97 %.